Validation of Noninvasive Body Sensor Network Technology in the Detection of Agitation in Dementia

Bankole, Azziza; Anderson, Martha; Smith‐Jackson, Tonya L.; Knight, Aubrey L.; Oh, Kyunghui; Brantley, Jeff S.; Barth, Adam T.; Lach, John

doi:10.1177/1533317512452036

Cited by 38 publications

(41 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[20], [26], blood volume pressure [20], heart rate [26], skin temperature [26], and EEG waveforms [15] have been used to train models of agitating behavior. Studies on the acceleration of the wrists, ankles, and waist are performed to understand their association with agitation scale variables [6], [30]. The downside of these techniques is that, they require an active engagement of the person while collecting data -which is highly inconvenient.…”

Section: False Alarmsmentioning

confidence: 99%

“…State-of-the-art agitation monitoring systems use two kinds of sensing techniques in general: invasive [26], [15], [6] and non-invasive [1], [12], [21]. Invasive techniques use physiological parameters such as EEG waveforms, skin temperature, skin conductance, pupil diameter, respiration rate, and accelerometer readings from wearable sensors to detect agitation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Kintense: A robust, accurate, real-time and evolving system for detecting aggressive actions from streaming 3D skeleton data

Nirjon

Greenwood

Torres

et al. 2014

2014 IEEE International Conference on Pervasive Computing and Communications (PerCom)

View full text Add to dashboard Cite

Abstract-Kintense is a robust, accurate, real-time, and evolving system for detecting aggressive actions such as hitting, kicking, pushing, and throwing from streaming 3D skeleton joint coordinates obtained from Kinect sensors. Kintense uses a combination of: (1) an array of supervised learners to recognize a predefined set of aggressive actions, (2) an unsupervised learner to discover new aggressive actions or refine existing actions, and (3) human feedback to reduce false alarms and to label potential aggressive actions. This paper describes the design and implementation of Kintense and provides empirical evidence that the system is 11% − 16% more accurate and 10% − 54% more robust to changes in distance, body orientation, speed, and person when compared to standard techniques such as dynamic time warping (DTW) and posture based gesture recognizers. We deploy Kintense in two multi-person households and demonstrate how it evolves to discover and learn unseen actions, achieves up to 90% accuracy, runs in real-time, and reduces false alarms with up to 13 times fewer user interactions than a typical system.

show abstract

Section: False Alarmsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Kintense: A robust, accurate, real-time and evolving system for detecting aggressive actions from streaming 3D skeleton data

Nirjon

Greenwood

Torres

et al. 2014

2014 IEEE International Conference on Pervasive Computing and Communications (PerCom)

View full text Add to dashboard Cite

show abstract

“…A pilot study was conducted with 6 elderly residents with dementia living in long-term care identified as being at high risk of agitated behaviors. 22 Nine hours of data were collected from each resident using devices applied at 3 sites while behaviors were simultaneously observed and annotated. The TEMPO data were processed using Teager energy analysis, 31 (Kaiser) which measures the mechanical energy of the movement, emphasizing the jerky and repetitive movements that tend to correspond to physical agitation.…”

Section: Theoretical Frameworkmentioning

confidence: 99%

Correlates Among Nocturnal Agitation, Sleep, and Urinary Incontinence in Dementia

Rose

Specht

Forch

2014

Am J Alzheimers Dis Other Demen

View full text Add to dashboard Cite

Family caregivers of elders with dementia often face the challenging behaviors of nighttime agitation, sleep disturbances, and urinary incontinence. To date, no study has examined the interrelationships of these behaviors in community-dwelling persons. This single group, descriptive study employs wireless body sensors to objectively collect data on nighttime agitation, sleep, and urinary incontinence in patients with dementia in their homes over a 5- to 7-day period. The aims are to (1) examine the feasibility and acceptability of the use of body sensors in community-dwelling persons with dementia; (2) describe patterns of nocturnal agitation, sleep continuity and duration, and nighttime urinary incontinence; and (3) examine the relationships among nocturnal agitation, sleep continuity and duration, and nighttime urinary incontinence. Data collection is in early stages and is still in progress. Challenges and advantages from preliminary data collection are reported.

show abstract

“…Physical agitation can be measured using a wrist-worn 3-axis accelerometer as demonstrated in [8] and discussed in the related work section. Verbal agitation requires a microphone close enough to the PWD to detect speech or other verbal outbursts.…”

Section: Sensor System Requirementsmentioning

confidence: 99%

“…The pilot study that was the precursor to the BESI project used accelerometers worn on the wrists, ankles, and waist of a PWD to assess agitation [8]. The study showed some correlation between CMAI scores and motion as measured using the Teager energy of the accelerometer data.…”

Section: B Related Work Detecting Agitationmentioning

confidence: 99%

Design of a Home Sensing System for Monitoring Agitation in People with Dementia with Real-World Deployment Considerations

Dugan¹

View full text Add to dashboard Cite

Alzheimer's disease affects a large number of elderly people throughout the world. Caring for a person with Alzheimer's dementia who lives at home rather than in a nursing facility can be extremely taxing on the family of the person with dementia (PWD). Caregivers report high levels of stress and depression, and that episodes of agitation are the most difficult part of caring for a PWD. An inability to adequately deal with agitation is the most important reason caregivers cite for moving the PWD to a nursing facility.The Behavioral and Environmental Sensing and Intervention (BESI) project aims to increase the caregivers' self-perceived ability to deal with agitation by predicting agitation events and providing the caregivers with real-time notifications and targeted interventions. Agitation is predicted based on the environmental conditions around the PWD measured continuously using a sensor system deployed in the home. Physical agitation and the behaviors that lead up agitation events are measured using a wearable inertial sensor on the PWD. The inertial and environmental data will be fed into a model personalized to each PWD that predicts agitation and suggests targeted interventions to the caregivers. This thesis presents the design of a sensor system for use in the BESI project that measures the environmental conditions in a house occupied by a caregiver and a PWD and detects agitation using a wearable accelerometer on the PWD. The system continuously collects sensor data from multiple rooms, and aggregates it in a central location for analysis. Designing home monitoring systems like this one, presents a number of significant engineering challenges that need to be addressed to collect, transmit, and synchronize data from many disparate sensing modalities. These challenges along with considerations for in-home deployment, such as system reliability, monitoring, repair, and appearance are discussed. The sensor system has been deployed in the homes' of two caregiver-PWD dyads with each deployment lasting seven days, and the results from these deployments are presented in terms of the amount of data collected from the environmental and body-worn sensors.

show abstract

Validation of Noninvasive Body Sensor Network Technology in the Detection of Agitation in Dementia

Abstract: The BSN shows promise from these pilot results. Further testing with a larger sample is needed to replicate these results.

Cited by 38 publications

References 33 publications

Kintense: A robust, accurate, real-time and evolving system for detecting aggressive actions from streaming 3D skeleton data

Kintense: A robust, accurate, real-time and evolving system for detecting aggressive actions from streaming 3D skeleton data

Correlates Among Nocturnal Agitation, Sleep, and Urinary Incontinence in Dementia

Design of a Home Sensing System for Monitoring Agitation in People with Dementia with Real-World Deployment Considerations

Contact Info

Product

Resources

About